JP4668421B2

JP4668421B2 - Method and apparatus for continuous analysis of cardiovascular activity of a subject

Info

Publication number: JP4668421B2
Application number: JP2000598065A
Authority: JP
Inventors: シュスターマン、ブラディミール; ロイトバルフ、アレックス; オルバッチ、ツビ
Original assignee: ウルトラシスインターナショナル（１９９３）リミテッド
Priority date: 1999-02-11
Filing date: 2000-02-11
Publication date: 2011-04-13
Anticipated expiration: 2020-02-11
Also published as: AU2569700A; DE60036080D1; ATE370701T1; JP2002536104A; IL128482A0; BR0008154B1; BR0008154A; ES2292427T3; EP1150604B1; EP1150604B8; EP1150604A1; ZA200106578B; AU769672B2; DE60036080T8; PT1150604E; IL128482A; DE60036080T2; WO2000047110A1

Abstract

A method for obtaining continuously and non-invasively one or more parameters relating to the cardiovascular system of a subject. The parameters obtainable by the method are systolic blood pressure, diastolic blood pressure, young modulus of an artery, cardiac output, relative changes in vascular resistance, and relative changes in vascular compliance. In accordance with the method, the ratio, kappa of the subject's blood flow velocity to the propagation speed of the subject's pulse wave is obtained continuously and non-invasively. kappa is then processed so as to obtain the instantaneous values of the desired parameters.

Description

【０００１】
（技術分野）
本発明は医療診断装置の分野に関し、特に、被験者の心臓血管活動を解析する装置に関する。
【０００２】
用語解説
以下に、ここで使用される用語をその略字と共に解説し、そのいくつかは標準であり他は新造語である。
プレチスモグラフ（PG）−血流を測定する装置。
脈通過時間（PTT）−動脈系統内の２点に脈圧ピークが到来する間に経過した時間、すなわちECG信号内の特定の点と動脈系統内の特定の点への連続する脈波の到来との間で経過した時間。
心拍出量（CO）−毎分心臓から大動脈内へ送出される血液量
血管伸展性（VCL）−圧力変化に対する血管量変化の割合。
ＡＲＥＡ−プレチスモグラフ信号のピークの下の面積
ピーク振幅（PA）−プレチスモグラフ信号のピーク振幅。
収縮期血圧（SP）−心臓周期の収縮期中の血圧。
拡張期血圧（DP）−心臓周期の弛緩期間中の血圧。
【０００３】
（背景技術）
例えば、異常に高いもしくは低い血圧が健康に対する重大な脅威となる人にとって、血圧および血管パラメータの連続的な非侵襲性監視が重要である。非侵襲性、連続的、血圧監視を行うためのいくつかの方法が開発されてきている。例えば、米国特許第４，４７５，５５４号にはオシロメータ測定値から血圧を求める装置が開示されている。これらの装置は肘上の腕や指の上に配置することができる膨張可能な腕帯を利用する。腕帯は下にある指の血管の内圧と平衡するまで膨張される。指の動脈内の血圧が変動すると、腕帯圧は帰還制御機構により血圧を平衡させるように調節される。任意の瞬間における血圧は腕帯圧に比例するものと考えられる。それは指の動脈の弾性および音が、実際には極端に変化するが、終始一定のままであるものと仮定している。そのため、これらの装置は長期間の血圧監視には向かない。さらに、定腕帯圧によりこれらの腕帯は患者にとって苦痛を与えるものとなり抹消血液循環に問題を生じることがある。したがって、これらのオシロメータ装置は連続的血圧監視に使用されることはめったにない。
【０００４】
プレチスモグラフ（PG）信号だけを解析して収縮期および拡張期血圧を推定するためのいくつかの研究が試みられてきている。しかしながら、これらの方法は高次の導関数を利用し、したがって被験者の動きにより実際には達成できない極端に低ノイズの信号を必要とする。さらに、数分にわたってデータを平均しなければならないため、これらの方法はリアルタイム血圧測定には使用できない。
【０００５】
オアの米国特許第４，８６９，２６２号、スラメックの米国特許第４，８０７，６３８号、およびスゴの米国特許第５，７０９，２１２号は、脈通過時間（PTT）だけから血圧を計算する装置を開示している。しかしながら、PTTだけから決定された血圧測定値の信頼度および再生度は正確な血圧測定値を与えるのに十分なものではない。
【０００６】
血圧のもう１つの非侵襲性監視方法がスミスの欧州特許出願ＥＰ０４４３２６７Ａ１に開示されている。この方法は収縮期および拡張期血圧の計算にPG信号およびPTTの両方を使用する。血管音および弾性率の変動が心拍数よりも低速であるものと仮定して、AC信号をDC信号で除すことによりPG信号を最初に各心臓周期内で正規化しなければならない。この正規化手順は経験的であり不正確である。さらに、収縮期および拡張期血圧の計算に使用される方程式も経験的であるため多くの場合不正確である。
【０００７】
心拍出量その他の血管特性（伸展性、抵抗、およびヤング率）の変化が血圧に影響を及ぼすことが前から知られている。さまざまな生理的過程がさまざまな原因による血圧変化を支配し、それが異なる原因から生じる場合には血圧の同じ変化に対して異なる内科療法が必要である。したがって、血圧変化の原因を決定することは治療を成功させるのに極めて重要である。しかしながら、従来技術のどの装置および方法にもこれらの要因の非侵襲性監視手段は開示されていない。さらに、従来技術の装置および方法は全て血圧に及ぼすこれらの要因の影響を無視している。
【０００８】
したがって、前記した従来技術の方法の欠点が実質的に低減されるもしくは解消される血圧、心拍出量、その他の血管特性の非侵襲性、連続的監視方法および装置が必要とされている。
【０００９】
（発明の開示）
本発明の文脈において、２つの明確に記載された、計算可能すなわち測定可能な変数は、２つの変数が互いに比例する時は、互いに同値であると考えられる。
【００１０】
以下の説明および特許請求の範囲において、ひとつの個体の圧脈波の伝播速度に対する血液流速の比率を示すのにκが使用される。
【００１１】
本発明は、拡張期および収縮期血圧をκを含む計算により求められるという新しい非自明な発見に基づいており、κが使用されない従来技術の方法により得られるものよりも正確である。
【００１２】
したがって、本発明はκの連続的非侵襲性測定方法および装置を含んでいる。本発明の好ましい実施例では、κは被験者のPG信号およびPTTから得られる。最も好ましい実施例では、κは全体が本開示の一部としてここに組み入れられている、例えば、ランドウエル．ディー．およびリフシツイー．エム．のStatistical Physics, Pergamon Press, 1979; ランドウエル．ディー．およびリフシツイー．エム．のFluid Mechanics, Pergamon Press, 1987; およびランドウエル．ディー．およびリフシツイー．エム．のTheory of Elasticity, Pergamon Press, 1986; およびカプランディーおよびグラスのUnderstanding Non-Linear Dynamics, Springer-Verlag. N.Y., 1995 に記載されている強不連続性の波動理論に従って被験者のPG信号およびPTTから得られる。さらに好ましい実施例では、κは次式で与え、
【数４２】
κ=1/(1/(PEAK・v)+1)
ここに、ｖはPTTに逆比例する脈波の伝播速度（脈波速度）であり、
【数４３】
PEAK=k₁・PTT・PA+k₂・AREA
であり、
ここに、PAおよびAREAはPG信号から得られる脈波のそれぞれ振幅および面積であり、k₁およびk₂は2つの経験的に得られる定数である。
【００１３】
もう１つの好ましい実施例では、κは
【数４４】

により与えられる。
【００１４】
κ測定値の精度を高めるために血管半径内の低速(0.01-0.05Hz)変動（血管運動音）を随意PG信号からフィルタリングにより除去することができる。それは、例えば、κの定義におけるPEAKをPEAK/(PEAKの低速成分)²で置換して実施することができる。PEAKの低速成分は、例えば、脈波の低域フィルタリングにより得ることができる。本発明の範囲内でκを連続的かつ非侵襲性に得る他の方法も考えられる。
【００１５】
被験者のPG信号を連続的に非侵襲的に得る手段が従来技術で知られており、例えば、光PGセンサとすることができる。本発明の範囲内で血管内の圧力波を測定する他の方法も考えられる。限定はしないが、それにはいくつかの光PG装置、インピーダンスPG装置、圧電、超音波、レーザー、その他のタイプのセンサの使用が含まれる。
【００１６】
PTTを連続的かつ非侵襲的に求める手段が従来技術で知られており、例えば、心電計モニタおよびPGセンサを含むことができる。この場合PTTはECG波内の特定の点、例えばＲピーク、とPGセンサへの対応する圧力波の到来との間の時間経過である。PTTを測定する他の手段は、例えば、同じ動脈血管に沿って互いに離されて皮膚に取り付けられる一対のPGセンサを含んでいる。この場合、PPTは２つの位置における圧力波の到来間の時間経過である。
【００１７】
したがって、本発明はさらに収縮期および拡張期血圧の連続的かつ非侵襲的測定値を得るようにκをリアルタイムで処理する装置を提供する。
【００１８】
さらに、本発明はヤング率、血管抵抗、心拍出量、および血管伸展性の連続的かつ非侵襲的測定値を得るようにκをリアルタイムで処理する装置を提供する。従来技術はこれらのパラメータを得るための方法を開示していない。
【００１９】
本発明により提供される拡張期および収縮期血圧、ヤング率、血管抵抗、心拍出量、および血管伸展性の測定値は、従来技術の装置により提供される測定値よりも、外部ノイズ、身体姿勢の変化、およびセンサ配置に対してよりロバストで影響されにくい。
【００２０】
本発明は、さらに、心拍出量の変化や血管伸展性の変化による被験者内の血圧の変化を示す指標を連続的かつ非侵襲的に得るようにκをリアルタイムで処理する装置を提供する。異なる生理的過程が原因の異なる血圧変化を支配し、異なる原因から生じる場合は血圧の同じ変化に対して異なる内科療法が必要であるため、本発明は適切な療法を決定するための手段を提供する。
【００２１】
好ましい実施例では、κは強い不連続性の波動理論に従って前記したパラメータを得るようにリアルタイムで処理される。最も好ましい実施例では、前記したパラメータは下記のアルゴリズム式を使用して得られる。
収縮期血圧（SP）
方法１
【数４５】
SP=ρv²Φ(κ,γ)
ここに、ρは血液密度、γは血液の熱力学ポアソン指数であり、
【数４６】

方法２
【数４７】
SP=(logv²)/α+2ρv²κ/3+λ,
ここに、λ=(log(2ρR/E₀h))/α，Ｒは動脈の半径、ｈは動脈壁の厚さ、E₀はゼロ圧に関するヤング率、αは経験的に得られる定数である。
方法３
【数４８】
SP=(logv²/(1-εH²))/α+2ρv²κ/3+λ,
ここに、εは経験的に得られる定数でありＨは心拍数である。
方法４
【数４９】
SP=［(logv²)/α+λ］/(1-κ),
方法５
【数５０】
SP=［(logv²/(1-εH²))/α+λ］/(1-κ),
拡張期血圧（DP）
【数５１】
DP=SP-ρv²κ,
ヤング率
方法１
【数５２】
E=(2R/h)(SP-DP)/κ,
方法２
【数５３】
E=(2R/h)SP/Φ(κ,γ),
方法３
【数５４】
E=(2R/h)・ρ・exp［(-λ+MP)α］,
ここに、MPは平均圧で、MP=(SP+2・DP)/3であり、SPもしくはDPはκを含むアルゴリズム式を使用して得られる。
方法４
【数５５】
E=(2R/h)・ρ・exp((-λ+SP・(1-κ))α),
心拍出量（CO）
【数５６】
CO=PEAK・｛v・［1+SP/(2ρ・v²)］｝²
ここに、SPはκを含むアルゴリズム式を使用して得られ、PEAKの低速成分は前記したフィルタリングにより除去される。
血管抵抗（VR）
【数５７】
VR=(SP-DP)/CO
ここに、SP,DP,およびCOのいずれか１つ以上がκを含むアルゴリズム式を使用して得られる。
血管伸展性（VC）
【数５８】
VC=PEAK/(SP-DP)
ここに、SP,およびDPのいずれか１つ以上がκを含む計算から得られる。本発明の範囲内でκから血管伸展性を得る他の方法も考えられる。
【００２２】
血圧へのVC,VRおよびCOの影響
SPの観察された変化へのCOの相対的寄与は、次式で定義されるパラメータINDEX1により与えられ、
【数５９】
INDEX1=∂SP/∂CO-∂SP/∂VC
ここに、SP,COおよびVCのいずれか１つ以上がκを含む計算から得られる。時間と共にINDEX1が増加することは主として心拍出量(CO)の変化によるSPの変化を示す。時間と共にINDEX1が減少することは主として血管伸展性(VC)の変化によるSPの変化を示す。
【００２３】
SPの観察された変化へのVRおよびCOの相対的寄与は、次式で定義されるパラメータINDEX2により与えられ、
【数６０】
INDEX2=∂SP/∂CO-∂SP/∂VR
ここに、SP,COおよびVCのいずれか１つ以上がκを含む計算から得られる。時間と共にINDEX2が増加することは主として心拍出量(CO)の変化によるSPの変化を示す。時間と共にINDEX2が減少することは主として血管抵抗(VR)の変化によるSPおよびDPの変化を示す。
【００２４】
（発明を実施するための最良の形態）
図１は本発明の好ましい実施例に従って監視される被験者１０を示す。連続的かつ非侵襲的に心電計を監視するために、被験者の胸部にECG電極１２が取り付けられている。連続的かつ非侵襲的に脈波を監視するために、被験者の指にPGセンサ１４が取り付けられている。ECG電極およびPGセンサからの信号が連続的にプロセッサ１６へ与えられる。プロセッサ１６はインターフェイス、Ａ／Ｄコンバータ、増幅器およびＰＣコンピュータ１８のシリアルポートへのケーブルを含んでいる。市販の血圧計２０を使用して任意の経験的に規定される定数を得るために、校正の目的で予備血圧測定が行われる。
【００２５】
プロセッサ１６により実施される処理の一般化されたフロー図が図２に示されている。ECGおよびPG信号はκの瞬時値を得るように、最初にリアルタイムで処理される。次に、κは所望するパラメータの瞬時値を得るようにリアルタイムで処理される。所望するパラメータの計算された値は格納および表示するためにＰＣ１８へリアルタイムで転送される。
【００２６】
例
次に、限定しない例により本発明を説明する。
【００２７】
方法
一群の１１人の被験者、年齢が２１−４４才の７人の男性および４人の女性、の血圧が本発明を使用して調べられた。１１人の被験者の内、１０人は正常な血圧を有することが判り、１人は高血圧に近かった。各被験者は少なくとも２回調べられた。各検査はほぼ１時間続き、次の姿勢における測定が含まれた、仰臥位（１５分）、座位（１５分）、および立位（１０分）。７人の被験者は自転車での管理された身体運動の１０分後もしくはヴァルサルヴァ試験中にも座位で測定が行われた。各被験者および各姿勢について別々にデータが処理された。
【００２８】
下記の装置の一方もしくは両方を使用して、各被験者から標準血圧測定値が得られた。
1. 市販の血圧測定装置（ＰＣコンピュータに接続された腕帯形圧力計を含むDynapulse 200M^TM）
2. 出力をＰＣコンピュータへ送信する装置(Ultramind)と組み合わされた指動脈からの連続振動血圧測定(Finapress^TM, Ohmeda)。
【００２９】
Finapress^TM装置が標準として使用された時は、血圧は連続的に測定されリアルタイムで保全された。Dynapulse^TM装置が使用された時は、検査中に個別の血圧測定が３−４回行われた。各検査の始めの標準血圧測定値が定数パラメータk₁,k₂を得るのに使用された。
【００３０】
各被験者からECGおよびPG信号が得られ、カスタムソフトウェアによりリアルタイムで処理された。処理には下記の連続操作が含まれた。
1. 平滑化（フィルタリングおよび高周波ノイズ）
2. PG信号のベースラインドリフト相関（1.0〜2.0Hzの遮断周波数を使用するハイパスフィルタリング）
3. ECGおよびPG信号のピーク認識手順を実施する。
4. PGピークの高さとしてPAを得る。
5. ECGピークと対応するPGピーク間の時間間隔としてPTTを計算する。
6. ECGピークからPGピークまでの時間間隔にわたるPG信号の積分によりAREAを計算する。
7. 心拍数を計算する。
8. 最尤原理に従ってカイ二乗判定法を使用して定数パラメータk₁,k₂，αおよびεを計算する。
【００３１】
次に、各被験者に対してSPおよびDPが次のように得られた。
各被験者からECGおよびPG信号が得られ、カスタムソフトウェアによりリアルタイムで処理された。処理には下記の連続操作が含まれた。
1. 平滑化（フィルタリングおよび高周波ノイズ）
2. PG信号のベースラインドリフト相関（1.0-2.0 Hzの遮断周波数を使用するハイパスフィルタリング）
3. CGおよびPG信号のピーク認識手順を実施する。
4. PG信号の高さとしてPAを得る。
5. ECGピークと対応するPGピーク間の時間間隔としてPTTを計算する。
6. ECGピークからPGピークまでの時間間隔にわたるPG信号の積分によりAREAを計算する。
7. 心拍数を計算する。
8. 本発明の方法に従ってSPおよびDPを計算する。
9. スミスの欧州特許出願ＥＰ０４４３２６７Ａ１の方法に従ってSPおよびDPを計算する。
【００３２】
必要に応じて検査中に時々定数パラメータが調節された。
【００３３】
結果が下記の方法で得られたものと比較された。
【００３４】
出力は下記の2つの部分からなっている。
1. 本発明に従ってかつスミスの方法に従って得られたSPおよびDP時系列。
2. SPおよびDP時系列と標準血圧測定値間の平均誤差およびroot-mean-square誤差。
【００３５】
結果
休息中の被験者に対して本発明に従って得られた血圧測定値、およびスミスの欧州特許出願ＥＰ０４４３２６７Ａ１の経験式により計算されたもの（表１）が標準装置により得られたものと比較された。本発明に従って得られたSPおよびDP測定値はスミスの方法により得られたものよりも安定であった。特に、本発明の方法３および５に従ってストレスの後で得られたSP測定値の平均誤差および標準偏差はスミスの方法により得られたものよりも、それぞれ、２および５倍小さかった。２６人の全被験者において、ＳＰ測定値が方法３および５に従って得られた時の平均誤差はスミスの方法により得られたものよりも１．６倍小さかった(p=0.023)。
【００３６】
表２は運動後に仰臥位もしくは座位にある被験者に対して得られた血圧測定値の結果を示す。本発明の５つの方法およびスミスの欧州特許出願ＥＰ０４４３２６７Ａ１の方法がFinapress^TMおよびDynapulse 200M^TMにより得られた測定値と比較された。特に、２６人の全被験者において方法３および５に従って得られたSP測定値間の平均誤差はスミスの得られた誤差の５４％であった(p=0.023)。
【表１】

表１．Finapress^TMもしくはDynapulse^TMを使用して得られた測定値と比較した本発明の５つの方法、およびスミスの方法により得られた血圧測定値間の平均誤差±標準偏差(mmHg)。
【表２】

表２．Finapress^TMおよびDynapulse 200^TMを使用して得られた測定値と比較した本発明の５つの方法、およびスミスの欧州特許出願ＥＰ０４４３２６７Ａ１の方法により得られた平均誤差±標準偏差(mmHg)。
【００３７】
ある程度の特異性で本発明が説明されてきたが、特許請求の範囲に明記された本発明の範囲および精神を逸脱することなくさまざまな通知および変更を行えることをお判り願いたい。
【００３８】
参照
米国特許
US 4,475,554 10/1984 ヒンドマン
US 4,807,638 2/1989 スラメック
US 4,869,262 9/1989 オア
US 5,709,212 1/98 スゴ
欧州特許出願
EP 0443267A1（US 484687に対応する） 2/1990 スミス
【図面の簡単な説明】
【図１】本発明の装置を使用する一実施例を示す図である。
【図２】本発明の一実施例に従った処理ステップの一般化されたフロー図である。[0001]
(Technical field)
The present invention relates to the field of medical diagnostic devices, and more particularly to a device for analyzing a subject's cardiovascular activity.
[0002]
Glossary Below, the terms used here are explained together with their abbreviations, some of which are standard and others are new terms.
Plethysmograph (PG)-A device that measures blood flow.
Pulse transit time (PTT)-The time that elapses between the arrival of pulse pressure peaks at two points in the arterial system, that is, the arrival of a continuous pulse wave at a specific point in the ECG signal and a specific point in the arterial system. Elapsed time between.
Cardiac output (CO)-the volume of blood delivered from the heart into the aorta per minute vascular extensibility (VCL)-the ratio of the change in blood vessel volume to the pressure change.
AREA—Area peak amplitude (PA) under the peak of the plethysmograph signal—Peak amplitude of the plethysmograph signal.
Systolic blood pressure (SP)-Blood pressure during systole of the cardiac cycle.
Diastolic blood pressure (DP)-Blood pressure during the relaxation period of the cardiac cycle.
[0003]
(Background technology)
For example, for people whose abnormally high or low blood pressure is a significant health threat, continuous non-invasive monitoring of blood pressure and vascular parameters is important. Several methods have been developed for non-invasive, continuous, blood pressure monitoring. For example, US Pat. No. 4,475,554 discloses an apparatus for determining blood pressure from oscillometer measurements. These devices utilize an inflatable armband that can be placed on the arm or finger on the elbow. The bracelet is inflated until it equilibrates with the internal pressure of the underlying finger blood vessel. When the blood pressure in the finger artery fluctuates, the brachial band pressure is adjusted by the feedback control mechanism to balance the blood pressure. The blood pressure at any moment is considered to be proportional to the armband pressure. It assumes that the elasticity and sound of the finger arteries actually vary drastically but remain constant throughout. Therefore, these devices are not suitable for long-term blood pressure monitoring. Furthermore, constant arm pressure can cause pain for the patient and can cause problems with peripheral blood circulation. Therefore, these oscillometer devices are rarely used for continuous blood pressure monitoring.
[0004]
Several studies have been attempted to analyze only the plethysmograph (PG) signal to estimate systolic and diastolic blood pressure. However, these methods utilize higher order derivatives and therefore require extremely low noise signals that cannot be achieved in practice by subject movement. Furthermore, these methods cannot be used for real-time blood pressure measurements because the data must be averaged over several minutes.
[0005]
Or, U.S. Pat. No. 4,869,262, Slamek, U.S. Pat. No. 4,807,638, and Sugo, U.S. Pat. No. 5,709,212 calculate blood pressure from pulse transit time (PTT) alone. An apparatus is disclosed. However, the reliability and reproducibility of blood pressure measurements determined solely from PTT is not sufficient to give accurate blood pressure measurements.
[0006]
Another non-invasive method of monitoring blood pressure is disclosed in Smith's European patent application EP 0443267 A1. This method uses both PG signals and PTT to calculate systolic and diastolic blood pressure. Assuming that vascular sound and elastic modulus variations are slower than heart rate, the PG signal must first be normalized within each heart cycle by dividing the AC signal by the DC signal. This normalization procedure is empirical and inaccurate. In addition, the equations used to calculate systolic and diastolic blood pressures are often inaccurate because they are also empirical.
[0007]
It has long been known that changes in cardiac output and other vascular properties (extensibility, resistance, and Young's modulus) affect blood pressure. Different physiological therapies dominate blood pressure changes due to different causes, and different medical therapies are needed for the same changes in blood pressure if they arise from different causes. Therefore, determining the cause of blood pressure changes is critical to successful treatment. However, none of the prior art devices and methods disclose non-invasive monitoring of these factors. Furthermore, all prior art devices and methods ignore the effects of these factors on blood pressure.
[0008]
Accordingly, there is a need for non-invasive, continuous monitoring methods and devices for blood pressure, cardiac output, and other vascular characteristics that substantially reduce or eliminate the disadvantages of the prior art methods described above.
[0009]
(Disclosure of the Invention)
In the context of the present invention, two clearly described, calculable or measurable variables are considered equivalent to each other when the two variables are proportional to each other.
[0010]
In the following description and claims, κ is used to indicate the ratio of blood flow velocity to the propagation velocity of a single individual's pressure pulse wave.
[0011]
The present invention is based on a new non-obvious discovery that diastolic and systolic blood pressures are determined by calculations involving κ and is more accurate than that obtained by prior art methods where κ is not used.
[0012]
Accordingly, the present invention includes a continuous non-invasive measurement method and apparatus for κ. In a preferred embodiment of the invention, κ is obtained from the subject's PG signal and PTT. In the most preferred embodiment, κ is incorporated herein in its entirety as part of the present disclosure, eg, Landwell. Dee. And Lifshits E. M. Statistical Physics, Pergamon Press, 1979; Landwell. Dee. And Lifshits E. M. Fluid Mechanics, Pergamon Press, 1987; and Landwell. Dee. And Lifshits E. M. Obtained from subject's PG signal and PTT according to the strong discontinuity wave theory described in Theory of Elasticity, Pergamon Press, 1986; and Kaplan Dee and Glass, Understanding Non-Linear Dynamics, Springer-Verlag. NY, 1995 It is done. In a further preferred embodiment, κ is given by:
[Expression 42]
κ = 1 / (1 / (PEAK ・ v) +1)
Here, v is the pulse wave velocity (pulse wave velocity) that is inversely proportional to PTT,
[Expression 43]
PEAK = k ₁ , PTT , PA + k ₂ , AREA
And
Here, PA and AREA are the amplitude and area of the pulse wave obtained from the PG signal, respectively, and k ₁ and k ₂ are two empirically obtained constants.
[0013]
In another preferred embodiment, κ is

Given by.
[0014]
In order to increase the accuracy of the κ measurement, slow (0.01-0.05 Hz) fluctuations (vascular motion sounds) within the blood vessel radius can be filtered out of the optional PG signal. This can be done, for example, by replacing PEAK in the definition of κ with PEAK / (slow component of PEAK) ² . The low-speed component of PEAK can be obtained by, for example, low-pass filtering of the pulse wave. Other methods of obtaining κ continuously and non-invasively within the scope of the present invention are also contemplated.
[0015]
Means for continuously and non-invasively obtaining a subject's PG signal are known in the prior art and can be, for example, an optical PG sensor. Other methods of measuring pressure waves in blood vessels within the scope of the present invention are also contemplated. This includes, but is not limited to, the use of several optical PG devices, impedance PG devices, piezoelectrics, ultrasound, lasers, and other types of sensors.
[0016]
Means for continuously and non-invasively determining PTT are known in the prior art and can include, for example, an electrocardiograph monitor and a PG sensor. In this case, PTT is the time lapse between a specific point in the ECG wave, for example the R peak, and the arrival of the corresponding pressure wave at the PG sensor. Other means of measuring PTT include, for example, a pair of PG sensors that are attached to the skin and separated from each other along the same arterial blood vessel. In this case, PPT is the time lapse between the arrival of pressure waves at the two locations.
[0017]
Thus, the present invention further provides an apparatus for processing κ in real time to obtain continuous and non-invasive measurements of systolic and diastolic blood pressure.
[0018]
Furthermore, the present invention provides an apparatus for processing κ in real time to obtain continuous and non-invasive measurements of Young's modulus, vascular resistance, cardiac output, and vascular extensibility. The prior art does not disclose a method for obtaining these parameters.
[0019]
The measurements of diastolic and systolic blood pressure, Young's modulus, vascular resistance, cardiac output, and vascular extensibility provided by the present invention are greater than those provided by prior art devices in terms of external noise, body Robust and less susceptible to changes in posture and sensor placement.
[0020]
The present invention further provides an apparatus for processing κ in real time so as to continuously and non-invasively obtain an index indicating a change in blood pressure in a subject due to a change in cardiac output or a change in vascular extensibility. Since different physiological processes dominate different blood pressure changes due to and different medical therapies are required for the same changes in blood pressure if they arise from different causes, the present invention provides a means to determine the appropriate therapy To do.
[0021]
In the preferred embodiment, κ is processed in real time to obtain the parameters described above according to strong discontinuity wave theory. In the most preferred embodiment, the aforementioned parameters are obtained using the following algorithm equation:
Systolic blood pressure (SP)
Method 1
[Equation 45]
SP = ρv ² Φ (κ, γ)
Where ρ is the blood density, γ is the thermodynamic Poisson index of the blood,
[Equation 46]

Method 2
[Equation 47]
SP = (logv ² ) / α + 2ρv ² κ / 3 + λ,
Where λ = (log (2ρR / E ₀ h)) / α, R is the radius of the artery, h is the thickness of the artery wall, E ₀ is the Young's modulus for zero pressure, and α is a constant obtained empirically. is there.
Method 3
[Formula 48]
SP = (logv ² / (1-εH ² )) / α + 2ρv ² κ / 3 + λ,
Here, ε is a constant obtained empirically, and H is the heart rate.
Method 4
[Formula 49]
SP = [(logv ² ) / α + λ] / (1-κ),
Method 5
[Equation 50]
SP = [(logv ² / (1-εH ² )) / α + λ] / (1-κ),
Diastolic blood pressure (DP)
[Formula 51]
DP = SP-ρv ² κ,
Young's modulus
Method 1
[Formula 52]
E = (2R / h) (SP-DP) / κ,
Method 2
[53]
E = (2R / h) SP / Φ (κ, γ),
Method 3
[Formula 54]
E = (2R / h) ・ ρ ・ exp [(-λ + MP) α],
Here, MP is an average pressure, MP = (SP + 2 · DP) / 3, and SP or DP is obtained using an algorithm formula including κ.
Method 4
[Expression 55]
E = (2R / h) ・ ρ ・ exp ((-λ + SP ・ (1-κ)) α),
Cardiac output (CO)
[Expression 56]
CO = PEAK ・ {v ・ [1 + SP / (2ρ ・ v ² )]} ²
Here, SP is obtained using an algorithm expression including κ, and the low-speed component of PEAK is removed by the above-described filtering.
Vascular resistance (VR)
[Equation 57]
VR = (SP-DP) / CO
Here, any one or more of SP, DP, and CO is obtained using an algorithm expression including κ.
Vascular extensibility (VC)
[Formula 58]
VC = PEAK / (SP-DP)
Here, one or more of SP and DP is obtained from the calculation including κ. Other methods of obtaining vascular extensibility from kappa are also contemplated within the scope of the present invention.
[0022]
Effects of VC, VR and CO on blood pressure
The relative contribution of CO to the observed change in SP is given by the parameter INDEX1 defined by
[Formula 59]
INDEX1 = ∂SP / ∂CO-∂SP / ∂VC
Here, any one or more of SP, CO and VC is obtained from the calculation including κ. An increase in INDEX1 with time indicates a change in SP mainly due to a change in cardiac output (CO). A decrease in INDEX1 with time indicates a change in SP mainly due to a change in vascular extensibility (VC).
[0023]
The relative contribution of VR and CO to the observed change in SP is given by the parameter INDEX2 defined by
[Expression 60]
INDEX2 = ∂SP / ∂CO-∂SP / ∂VR
Here, any one or more of SP, CO and VC is obtained from the calculation including κ. An increase in INDEX2 with time indicates changes in SP mainly due to changes in cardiac output (CO). A decrease in INDEX2 with time indicates changes in SP and DP mainly due to changes in vascular resistance (VR).
[0024]
(Best Mode for Carrying Out the Invention)
FIG. 1 shows a subject 10 being monitored according to a preferred embodiment of the present invention. An ECG electrode 12 is attached to the subject's chest for continuous and non-invasive monitoring of the electrocardiograph. A PG sensor 14 is attached to the subject's finger to continuously and non-invasively monitor the pulse wave. Signals from the ECG electrode and the PG sensor are continuously provided to the processor 16. The processor 16 includes an interface, an A / D converter, an amplifier, and a cable to the serial port of the PC computer 18. In order to obtain any empirically defined constant using a commercially available sphygmomanometer 20, preliminary blood pressure measurements are made for calibration purposes.
[0025]
A generalized flow diagram of the processing performed by the processor 16 is shown in FIG. ECG and PG signals are first processed in real time to obtain an instantaneous value of κ. Κ is then processed in real time to obtain the instantaneous value of the desired parameter. The calculated values of the desired parameters are transferred in real time to the PC 18 for storage and display.
[0026]
Examples The invention will now be illustrated by non-limiting examples.
[0027]
Methods The blood pressure of a group of 11 subjects, 7 men aged 21-44 and 4 women, was examined using the present invention. Of the 11 subjects, 10 were found to have normal blood pressure and one was close to hypertension. Each subject was examined at least twice. Each exam lasted approximately 1 hour and included measurements in the following postures: supine (15 minutes), sitting (15 minutes), and standing (10 minutes). Seven subjects were also measured in the sitting position 10 minutes after controlled physical exercise on the bicycle or during the Valsalva test. Data was processed separately for each subject and each posture.
[0028]
Standard blood pressure measurements were obtained from each subject using one or both of the following devices.
1. A commercially available blood pressure measurement device (Dynapulse 200M ^TM including an armband pressure gauge connected to a PC computer)
2. Continuous vibration blood pressure measurement from finger arteries (Finapress ^™ , Ohmeda) combined with a device (Ultramind) that sends output to a PC computer.
[0029]
When the Finapress ^™ device was used as a standard, blood pressure was continuously measured and maintained in real time. When the Dynapulse ^™ device was used, individual blood pressure measurements were taken 3-4 times during the exam. The standard blood pressure measurement at the beginning of each test was used to obtain the constant parameters k ₁ and k ₂ .
[0030]
ECG and PG signals were obtained from each subject and processed in real time by custom software. The treatment included the following sequential operations.
1. Smoothing (filtering and high frequency noise)
2. PG signal baseline drift correlation (high-pass filtering using a cut-off frequency of 1.0 to 2.0 Hz)
3. Perform peak recognition procedure for ECG and PG signals.
4. Get PA as the height of the PG peak.
5. Calculate PTT as the time interval between the ECG peak and the corresponding PG peak.
6. Calculate AREA by integrating the PG signal over the time interval from ECG peak to PG peak.
7. Calculate heart rate.
8. Calculate the constant parameters k ₁ , k ₂ , α, and ε using the chi-square decision method according to the maximum likelihood principle.
[0031]
Next, SP and DP were obtained for each subject as follows.
ECG and PG signals were obtained from each subject and processed in real time by custom software. The treatment included the following sequential operations.
1. Smoothing (filtering and high frequency noise)
2. PG signal baseline drift correlation (high-pass filtering using a cut-off frequency of 1.0-2.0 Hz)
3. Perform peak recognition procedure for CG and PG signals.
4. Get PA as the height of the PG signal.
5. Calculate PTT as the time interval between the ECG peak and the corresponding PG peak.
6. Calculate AREA by integrating the PG signal over the time interval from ECG peak to PG peak.
7. Calculate heart rate.
8. Calculate SP and DP according to the method of the present invention.
9. Calculate SP and DP according to the method of Smith's European Patent Application EP 0 443 267 A1.
[0032]
Constant parameters were adjusted from time to time during testing as needed.
[0033]
The results were compared with those obtained by the following method.
[0034]
The output consists of two parts:
1. SP and DP time series obtained according to the invention and according to Smith's method.
2. Mean and root-mean-square error between SP and DP time series and standard blood pressure measurements.
[0035]
Results Blood pressure measurements obtained in accordance with the present invention for resting subjects and those calculated by the empirical formula of Smith's European patent application EP 0443267A1 (Table 1) were compared with those obtained by the standard apparatus. The SP and DP measurements obtained according to the present invention were more stable than those obtained by the Smith method. In particular, the mean error and standard deviation of the SP measurements obtained after stress according to methods 3 and 5 of the present invention were 2 and 5 times smaller than those obtained by the Smith method, respectively. In all 26 subjects, the mean error when SP measurements were obtained according to methods 3 and 5 was 1.6 times smaller than that obtained by the Smith method (p = 0.023).
[0036]
Table 2 shows the blood pressure measurement results obtained for subjects in supine or sitting position after exercise. The five methods of the present invention and the method of Smith's European patent application EP 0443267 A1 were compared with the measurements obtained with Finapress ^™ and Dynapulse 200M ^™ . In particular, the average error between SP measurements obtained according to methods 3 and 5 in all 26 subjects was 54% of the error obtained for Smith (p = 0.023).
[Table 1]

Table 1. Mean error ± standard deviation (mmHg) between the five methods of the present invention compared to the measurements obtained using Finapress ^™ or Dynapulse ^™ and the blood pressure measurements obtained by the Smith method.
[Table 2]

Table 2. Mean error ± standard deviation (mmHg) obtained by the five methods of the present invention compared to measurements obtained using Finapress ^™ and Dynapulse 200 ^™ , and the method of Smith's European Patent Application EP 0443267 A1.
[0037]
Although the invention has been described with some specificity, it should be understood that various notices and modifications can be made without departing from the scope and spirit of the invention as set forth in the claims.
[0038]
Reference US Patent
US 4,475,554 10/1984 Hindman
US 4,807,638 2/1989 Slamek
US 4,869,262 9/1989 Or
US 5,709,212 1/98 Sugo European patent application
EP 0443267A1 (corresponds to US 484687) 2/1990 Smith [Brief description of the drawings]
FIG. 1 shows an embodiment using the apparatus of the present invention.
FIG. 2 is a generalized flow diagram of processing steps according to one embodiment of the present invention.

Claims

A method for continuously and non-invasively obtaining at least two of the following parameters i to vi of a subject,
i. Systolic blood pressure (SP),
ii. Diastolic blood pressure (DP),
iii. Young's modulus of the artery,
iv. Cardiac output (CO),
v. Relative change in vascular resistance (VR),
vi. Relative change in vascular extensibility (VC),
The method
(A) obtaining a ratio of the subject's blood flow velocity to the pulse wave velocity of the subject, κ, substantially continuously and non-invasively for each beat;
(B) processing κ substantially in real time every beat to obtain the desired instantaneous values of the at least two parameters;
Including methods.

The method according to claim 1, wherein κ is obtained by continuously and non-invasively processing a PG (plethysmograph) signal and PTT (pulse transit time) obtained from a subject.

The method of claim 2, wherein κ is obtained according to the following algorithm equation:
[Expression 1]
κ = 1 / (1 / (PEAK ・ v) +1),
Where v is the pulse velocity, and
[Expression 2]
PEAK = k ₁・ PTT ・ PA + k ₂・ AREA,
Where PA and AREA are the amplitude and area of the pulse wave obtained from the PG signal, respectively, and k ₁ and k ₂ are empirically obtained methods.

The method of claim 2, wherein κ is obtained according to the following algorithm equation:

Where PA is the pulse wave amplitude obtained from the PG signal.

5. A method as claimed in claim 3 or 4, further comprising the step of filtering slow fluctuations in the pulse wave.

6. The method according to claim 5, wherein low-speed fluctuations in PEAK are filtered by replacing PEAK of claim 3 with PEAK / (low-speed component of PEAK) ² .

The method of claim 1, wherein the processing required to calculate the subject's systolic blood pressure in step (a) comprises the following algorithm equation:
[Expression 4]
SP = ρv ² Φ (κ, γ),
Where ρ is the blood density, γ is the thermodynamic Poisson index of blood, v is the pulse wave velocity, and

The way that is.

The method of claim 1, wherein the processing required to calculate the subject's systolic blood pressure in step (a) comprises the following algorithm equation:
[Formula 6]
SP = (logv ² ) / α + 2ρv ² κ / 3 + λ,
Where ρ is the blood density, v is the pulse wave velocity, and λ = (log (2ρR / E ₀ h)) / α, where R is the radius of the artery, E ₀ is the Young's modulus for zero pressure, h is the thickness of the artery wall and α is an empirically obtained method.

The method of claim 1, wherein the processing required to calculate the subject's systolic blood pressure in step (a) comprises the following algorithm equation:
[Expression 7]
SP = [(logv ² ) / α + λ] / (1-κ),
Here, v is the pulse wave velocity, and λ = (log (2ρR / E ₀ h)) / α, R is the radius of the artery, E ₀ is the Young's modulus for zero pressure, ρ is the blood density, and h is the artery The wall thickness, α is an empirically obtained method.

The method of claim 1, wherein the processing required to calculate the subject's systolic blood pressure in step (a) comprises the following algorithm equation:
[Equation 8]
SP = (logv ² / (1-εH ² )) / α + 2ρv ² κ / 3 + λ,
Where ρ is the blood density, v is the pulse wave velocity, H is the heart rate, and λ = (log (2ρR / E ₀ h)) / α, where R is the radius of the artery and E ₀ is the zero pressure. Young's modulus with respect to h, h is the thickness of the arterial wall, and ε and α are obtained empirically.

The method of claim 1, wherein the processing required to calculate the subject's systolic blood pressure in step (a) comprises the following algorithm equation:
[Equation 9]
SP = [(logv ² / (1-εH ² )) / α + λ] / (1-κ),
Where v is the pulse wave velocity, H is the heart rate, and λ = (log (2ρR / E ₀ h)) / α, where R is the radius of the artery, E ₀ is the Young's modulus for zero pressure, h Is the thickness of the arterial wall, ρ is the blood density, ε and α are empirically obtained methods.

The method of claim 1, wherein the processing required to calculate the subject's diastolic blood pressure in step (a) comprises the following algorithm equation:
[Expression 10]
DP = SP-ρv ² κ,
Here, SP is systolic blood pressure, ρ is blood density, and v is a pulse wave velocity.

2. The method of claim 1, wherein the processing required to calculate the arterial Young's modulus of the subject in step (a) includes the following algorithm equation:
## EQU11 ##
E = (2R / h) (SP-DP) / κ,
Where R is the radius of the artery, h is the thickness of the artery wall, SP is the systolic blood pressure, and DP is the diastolic blood pressure.

2. The method of claim 1, wherein the processing required to calculate the arterial Young's modulus of the subject in step (a) includes the following algorithm equation:
[Expression 12]
E = (2R / h) SP / Φ (κ, γ),
Where R is the radius of the artery, h is the thickness of the artery wall, SP is the systolic blood pressure, γ is the thermodynamic Poisson index of the blood,

The way that is.

15. The method of claim 14, further comprising the step of filtering slow fluctuations in the pulse wave .

2. The method of claim 1, wherein the processing required to calculate the arterial Young's modulus of the subject in step (a) includes the following algorithm equation:
[Expression 14]
E = (2R / h) ・ ρ ・ exp (-λ + MP) α,
Where R is the radius of the artery, h is the thickness of the artery wall, ρ is the blood density, MP = (SP + 2 · DP) / 3, where SP is the systolic blood pressure and DP is the diastolic blood pressure And at least one of systolic blood pressure or diastolic blood pressure is obtained using an algorithmic expression including κ, and λ = (log (2ρR / E ₀ h)) / α, where E ₀ is The Young's modulus for zero pressure, α is a constant obtained empirically.

2. The method of claim 1, wherein the processing required to calculate the arterial Young's modulus of the subject in step (a) includes the following algorithm equation:
[Expression 15]
E = (2R / h) ・ ρ ・ exp ((-λ + SP ・ (1-κ)) α),
Where R is the radius of the artery, h is the thickness of the artery wall, ρ is the blood density, SP is the systolic blood pressure, and λ = (log (2ρR / E ₀ h)) / α, E ₀ is the Young's modulus for zero pressure and α is a constant obtained empirically.

The method of claim 1, wherein the processing required to calculate the subject's cardiac output in step (a) comprises the following algorithmic expression:
[Expression 16]
CO = PEAK ・ {v ・ [1 + SP / (2ρ ・ v ² )]} ²
Where SP is the systolic blood pressure obtained using an algorithmic expression including κ, ρ is the blood density, v is the pulse wave velocity, and
[Expression 17]
PEAK = k ₁ ・ PTT ・ PA + k ₂ ・ AREA,
Where PA and AREA are the amplitude and area of the pulse wave peak obtained from the PG signal, respectively, and k ₁ and k ₂ are obtained empirically .

The method of claim 1, wherein the processing required to calculate the subject's cardiac resistance in step (a) comprises the following algorithm equation:
[Formula 18]
VR = (SP-DP) / CO
Here, one or more of SP, DP and CO is obtained from a calculation including κ.

The method of claim 1, wherein the processing required to calculate the subject's vascular extensibility in step (a) comprises the following algorithm equation:
[Equation 19]
VC = PEAK / (SP-DP)
here,
PEAK = k ₁・ PTT ・ PA + k ₂・ AREA,
Where PA and AREA are the amplitude and area of the pulse wave obtained from the PG signal, respectively, and k ₁ and k ₂ are empirically obtained methods.

A method for continuously and non-invasively determining whether a change in blood pressure of a subject is due to a change in cardiac output or a change in vascular extensibility, the method comprising:
(A) the ratio of the subject's blood flow velocity to the pulse wave velocity of the subject, obtaining κ substantially continuously and non-invasively for each beat;
(B) processing κ substantially in real time every beat to obtain instantaneous values of the subject's systolic blood pressure SP, cardiac output CO and vascular extensibility VC;
(C) processing the subject's SP, CO, and VC in real time to obtain an instantaneous value of the following algorithmic expression:
[Expression 20]
INDEX1 = ∂SP / ∂CO-∂SP / ∂VC
INDEX1, which increases with time, indicates a change in the subject's blood pressure due to a change in cardiac output, otherwise it is determined that the change in the subject's blood pressure is due to a change in vascular extensibility.

A method for continuously and non-invasively determining whether a change in a subject's blood pressure is due to a change in a subject's cardiac output or a change in vascular resistance, the method comprising:
(A) the ratio of the subject's blood flow velocity to the pulse wave velocity of the subject, obtaining κ substantially continuously and non-invasively for each beat;
(B) processing κ in substantially real time every beat to obtain instantaneous values of the subject's systolic blood pressure SP, cardiac output CO and vascular extensibility VR;
(C) processing the subject's SP, CO and VR in real time to obtain an instantaneous value of the following algorithmic expression:
[Expression 21]
INDEX2 = ∂SP / ∂CO-∂SP / ∂VR
INDEX2, which increases with time, indicates a change in the subject's blood pressure due to a change in cardiac output, otherwise it is determined that the change in the subject's blood pressure is due to a change in vascular resistance.

A device for continuously and non-invasively obtaining at least two of the following parameters i to vi of a subject (10):
i. Systolic blood pressure (SP),
ii. Diastolic blood pressure (DP),
iii. Young's modulus of the artery,
iv. Cardiac output (CO),
v. Relative change in vascular resistance (VR),
vi. Relative change in vascular extensibility (VC),
The device
(A) the ratio of the blood flow velocity of the subject to the propagation velocity of the pulse wave of the subject, a device that obtains κ substantially continuously and non-invasively for each beat;
(B) a device (16) for processing κ substantially in real time for each beat so as to obtain an instantaneous value of a desired parameter;
Including the device.

24. The apparatus according to claim 23, wherein κ is obtained by continuously and non-invasively processing a PG (plethysmograph) signal and PTT (pulse transit time) obtained from a subject (10).

25. The apparatus of claim 24, wherein κ is obtained according to the following algorithm equation:
[Expression 22]
κ = 1 / (1 / (PEAK ・ v) +1),
Where v is inversely proportional to PTT, and
[Expression 23]
PEAK = k ₁・ PTT ・ PA + k ₂・ AREA,
Where PA and AREA are the amplitude and area of the pulse wave obtained from the PG signal, and k ₁ and k ₂ are empirically obtained devices.

26. The apparatus of claim 25, wherein the apparatus is capable of filtering low speed fluctuations in the pulse wave .

27. The apparatus according to claim 26, wherein low speed fluctuations in the pulse wave are filtered by replacing PEAK of claim 25 with PEAK / (low speed component of PEAK) ² .

24. The apparatus of claim 23, wherein κ is obtained according to the following algorithm equation:
[Expression 24]
κ = 1 / ((1 / PA) +1)
Here, PA is a device that is the amplitude of the pulse wave obtained from the PG signal .

24. The apparatus of claim 23, wherein the processing required to calculate the subject's systolic blood pressure in step (a) comprises the following algorithm equation:
[Expression 25]
SP = ρv ² Φ (κ, γ),
Where ρ is the blood density, γ is the thermodynamic Poisson index of blood, v is the pulse wave velocity, and

Is a device.

24. The apparatus of claim 23, wherein the processing required to calculate the subject's systolic blood pressure in step (a) comprises the following algorithm equation:
[Expression 27]
SP = (logv ² ) / α + 2ρv ² κ / 3 + λ,
Where ρ is the blood density and λ = (log (2ρR / E ₀ h)) / α, where R is the radius of the artery, E ₀ is the Young's modulus for zero pressure, and h is the thickness of the artery wall. That is, α is an empirically obtained device.

24. The apparatus of claim 23, wherein the processing required to calculate the subject's systolic blood pressure in step (a) comprises the following algorithm equation:
[Expression 28]
SP = [(logv ² ) / α + λ] / (1-κ),
Here, v is the pulse wave velocity, and λ = (log (2ρR / E ₀ h)) / α, R is the radius of the artery, E ₀ is the Young's modulus for zero pressure, ρ is the blood density, and h is the artery The thickness of the wall, α is an empirically obtained device.

24. The apparatus of claim 23, wherein the processing required to calculate the subject's systolic blood pressure in step (a) comprises the following algorithm equation:
[Expression 29]
SP = (logv ² / (1-εH ² )) / α + 2ρv ² κ / 3 + λ,
Where ρ is the blood density, v is the pulse wave velocity, H is the cardiac output, and λ = (log (2ρR / E ₀ h)) / α, where R is the radius of the artery and E ₀ is Young's modulus for zero pressure, h is the arterial wall thickness, and ε and α are empirically obtained devices.

24. The apparatus of claim 23, wherein the processing required to calculate the subject's systolic blood pressure in step (a) comprises the following algorithm equation:
[30]
SP = [(logv ² / (1-εH ² )) / α + λ] / (1-κ),
Where v is the pulse wave velocity, H is the heart rate, and λ = (log (2ρR / E ₀ h)) / α, where R is the radius of the artery, E ₀ is the Young's modulus for zero pressure, h Is the thickness of the artery wall, ρ is the blood density, and ε and α are empirically obtained devices.

24. The apparatus of claim 23, wherein the processing required to calculate a subject's diastolic blood pressure in step (a) comprises the following algorithm equation:
[31]
DP = SP-ρv ² κ,
Here, SP is the systolic blood pressure, ρ is the blood density, and v is the pulse wave velocity.

24. The apparatus of claim 23, wherein the processing required to calculate the subject's arterial Young's modulus in step (a) comprises the following algorithmic expression:
[Expression 32]
E = (2R / h) (SP-DP) / κ,
Where R is the radius of the artery, h is the thickness of the artery wall, SP is the systolic blood pressure, and DP is the diastolic blood pressure.

24. The apparatus of claim 23, wherein the processing required to calculate the subject's arterial Young's modulus in step (a) comprises the following algorithmic expression:
[Expression 33]
E = (2R / h) SP / Φ (κ, γ),
Where R is the radius of the artery, h is the thickness of the artery wall, SP is the systolic blood pressure, γ is the thermodynamic Poisson index of the blood,

Is a device.

24. The apparatus of claim 23, wherein the processing required to calculate the subject's arterial Young's modulus in step (a) comprises the following algorithmic expression:
[Expression 35]
E = (2R / h) ・ ρ ・ exp (-λ + MP) α,
Where R is the radius of the artery, h is the thickness of the artery wall, ρ is the blood density, MP = (SP + 2 · DP) / 3, where SP is the systolic blood pressure and DP is the diastolic blood pressure And at least one of systolic blood pressure or diastolic blood pressure is obtained using an algorithmic expression including κ, and λ = (log (2ρR / E ₀ h)) / α, where E ₀ is A device where Young's modulus for zero pressure, α is a constant obtained empirically.

36. The apparatus of claim 35, wherein the processing required to calculate the subject's arterial Young's modulus in step (a) comprises the following algorithmic expression:
[Expression 36]
E = (2R / h) ・ ρ ・ exp ((-λ + SP ・ (1-κ)) α
Where R is the radius of the artery, h is the thickness of the artery wall, ρ is the blood density, SP is the systolic blood pressure, and λ = (log (2ρR / E ₀ h)) / α, E ₀ is the Young's modulus for zero pressure and α is a constant obtained empirically.

24. The apparatus of claim 23, wherein the processing required to calculate the subject's cardiac output in step (a) comprises the following algorithmic expression:
[Expression 37]
CO = PEAK ・ {v ・ [1 + SP / (2ρ ・ v ² )]} ²
Where SP is the systolic blood pressure obtained using an algorithmic expression including κ, and
[Formula 38]
PEAK = k ₁・ PTT ・ PA + k ₂・ AREA,
Where PA and AREA are the amplitude and area of the pulse wave obtained from the PG signal, and k ₁ and k ₂ are empirically obtained devices.

24. The apparatus of claim 23, wherein the processing required to calculate the subject's cardiac resistance in step (a) comprises the following algorithmic expression:
[39]
VR = (SP-DP) / CO
Here, an apparatus obtained from a calculation in which any one or more of SP, DP and CO includes κ.

A device for continuously and non-invasively determining whether a change in blood pressure of a subject is due to a change in cardiac output or a change in vascular extensibility,
(A) the ratio of the blood flow velocity of the subject to the propagation velocity of the pulse wave of the subject, a device that obtains κ substantially continuously and non-invasively for each beat;
(B) a device (16) for processing κ substantially in real time for each beat so as to obtain instantaneous values of a subject's systolic blood pressure SP, cardiac output CO and vascular extensibility VR;
(C) a device for processing a subject's SP, CO and VC in real time so as to obtain an instantaneous value of the following algorithm formula,
[Formula 40]
INDEX1 = ∂SP / ∂CO-∂SP / ∂VC
INDEX1, which increases with time, indicates a change in the subject's blood pressure due to a change in cardiac output, otherwise the change in the subject's blood pressure is caused by a change in vascular extensibility.

A device for continuously and non-invasively determining whether a change in a subject's blood pressure is due to a change in a subject's vascular resistance, the device comprising:
(A) the ratio of the blood flow velocity of the subject to the propagation velocity of the pulse wave of the subject, a device that obtains κ substantially continuously and non-invasively for each beat;
(B) a device (16) for processing κ substantially in real time for each beat so as to obtain instantaneous values of a subject's systolic blood pressure SP, cardiac output CO and vascular extensibility VR;
(C) a device for processing the subject's SP, CO and VR in real time so as to obtain an instantaneous value of the following algorithm formula,
[Expression 41]
INDEX2 = ∂SP / ∂CO-∂SP / ∂VR
INDEX2, which increases with time, indicates a change in the subject's blood pressure due to a change in cardiac output, otherwise the change in the subject's blood pressure is attributed to a change in vascular resistance.